The present invention relates to an apparatus and a method of video signal processing. Particularly, this invention relates to an apparatus and a method of video signal processing for matrix display apparatus, such as, plasm display panels (PDPs), field emission displays (FEDs), digital micromirror devices (DMDs) and electroluminescent displays (ELs).
Pictures are displayed on PDPs, ELs and FEDs at a particular number of levels of gradation of color. The number is limited under digital processing for PDPs that display pictures with division of one field to sub-fields, and ELs and FEDS that display pictures with pulse width modulation (PWM).
PDPs, ELs and FEDs require reverse-gamma correction of video signals that have undergone gamma correction for regaining linear gradation.
Matrix display apparatus execute multi-gradation processing using dither matrices for achieving continuous gradation that would be degraded due to digital reverse-gamma correction.
Multi-gradation processing use dither matrices each having groups of adjacent pixel dots to provide intermediate gradation of the gradation that would otherwise be degraded due to reverse-gamma correction.
Six bit-gradation display apparatus display pictures with upper significant six bits of 8-bit dot data by means of dither matrices each having adjacent 2×2 dots on which a noise pattern is superimposed for the low significant two bits that will overflow in each dither matrix. These processing provide pictures with 8-bit gradation under visual integration.
Illustrated in FIG. 1 are well known dither coefficients matrix patterns. Shown here are dither coefficients patterns, each of which has a matrix of 2×2 dots (a, b, c and d). Each dot corresponds to one pixel of red (R), green (G) and blue (B).
The dither coefficients patterns are added to video signals for dot data on a display panel, such as a PDP. In detail, dither coefficients (a, b, a, b, . . . ) and (c, d, c, d, . . . ) are added to luminance values of R, G, and B of dot data on odd and even lines, respectively, of a dot matrix on the PDP. The addition starts from the head dot on each odd or even line. Furthermore, the dither coefficients (a, b, c and d) are added to the video signals for four adjacent dots of the same color for each of R, G and B.
Illustrated in (A) of FIG. 1 are two dither coefficients patterns (1) and (2) where the coefficients for the pattern (a, b, c, and d) are (0, 1, 2, and 3) and (3, 2, 1 and 0), respectively. The two patterns are switched for each field.
Illustrated in (B) of FIG. 1 are four dither coefficients patterns (1) to (4) where the coefficients for the patterns (a, b, c, and d) are (0, 1, 2, and 3), (2, 0, 3 and 1), (3, 2, 1, and 0) and (1, 3, 0 and 2), respectively. The four patterns are switched cyclically for each field.
Illustrated in FIG. 2 is video signal processing using the two dither coefficients patterns (1) and (2) shown in (A) of FIG. 1.
In detail, the dither coefficients (0, 1, 2, and 3) of the pattern (1) are added to dot data (9, 17, 3 and 5) of an input 8-bit video signal shown in (A) of FIG. 2 to obtain data (9, 18, 5 and 8).
The values (9, 18, 5 and 8) are larger than the 8-bit values (9, 17, 3 and 5) of the input video signal. The values (9, 18, 5 and 8) are then processed by limiting the values that overflow the input 8-bit values and also dropping the values of lower two bits to obtain a 6-bit output video signal of data (8, 16, 4 and 8).
Each data of (8, 16, 4 and 8) is expressed with a multiple number of four. The actual data for the 6-bit output video signal are (2, 4, 1 and 2). The 6-bit output video signal is a multi-gradation signal for which the number of gradation levels appear to be increased to eight bits for 6-bit PDPs with the dither coefficients pattern (1).
The data (9, 18, 5 and 8) may be processed by limiting the values that overflow the input 8-bit values (9, 17, 3 and 5) for 8-bit PDPs to provide 10-bit multi-gradation without dropping low tow-bit values.
Next, the dither coefficients (3, 2, 1, and 0) of the pattern (2) are added to dot data (9, 17, 3 and 5) of an input 8-bit video signal shown in (B) of FIG. 2 to obtain data (12, 19, 4 and 5).
The values (12, 19, 4 and 5) are also larger than the 8-bit values (9, 17, 3 and 5) of the input video signal. The values (12, 19, 4 and 5) are processed by limiting the values that overflow the input 8-bit values and dropping the values of lower two bits to obtain a 6-bit output video signal of data (12, 16, 4 and 4).
Each of the data (12, 16, 4 and 4) is also expressed with a multiple number of four. The actual data for the 6-bit output video signal are (3, 4, 1 and 1). The 6-bit output video signal is a multi-gradation signal for which the number of gradation levels appear to be increased to eight bits for 6-bit PDPs with the dither coefficients pattern (2).
The output video signals shown in (A) and (B) of FIG. 2 are switched for each field.
Video signal processing using the dither coefficients patterns (1) to (4) shown in (B) of FIG. 1 are basically the same as those discussed above with reference to FIG. 2. The dither coefficients patterns (1) to (4) are switched cyclically for each field to provide multi-gradation spatially continuous than those provided by the dither coefficients patterns (1) and (2) shown in (A) of FIG. 1.
Video signals undergo digital reverse-gamma correction to provide linear gradation before being supplied to display apparatus such as PDPs. The reverse-gamma correction decreases the number of gradation levels at low luminance level to cause uncontinuous gradation, thus resulting degradation of pictures.
PDPs provides gradation of color by constituting one field with sub-fields with different luminance weighting and selecting some of the sub-fields. The sub-field selection sometimes causes differences in visual luminance between adjacent gradations. This results in degradation of still and moving pictures with pseudo edges generated on pictures.
The dither coefficients patterns discussed above are employed for pictures of liner gradation which would otherwise be degraded due to generation of pseudo edges on pictures. The dither coefficients patterns shown in FIG. 1 are however applied for all the gradations with the same dither coefficients.
These dither coefficients patterns contribute degradation of pictures with pseudo edges due to a big luminance difference between adjacent gradations for which the dither coefficients give a big difference in the number of sub-fields selected at intermediate to high luminance levels.
Degradation of gradation due to digital reverse-gamma correction differs over low to high luminance levels. There is a big difference in such degradation particularly at the low luminance level.
The dither coefficients patterns shown in FIG. 1 are applied for all the gradations with the same dither coefficients as discussed above. This results in continuity only for a part of gradations on pictures.